Quantifying-Character-Similarity

This repo includes

Tutorials: notebook
Homoglyphs Dictionaries: Simplified Chinese, Traditional Chinese, Japanese, Korean.
CJK Fonts are here.
Scripts for japanese supply chain matching
Scripts for Synthetic Chinese, Japanese, Korean placenames matching
Scripts for ViT training and Inference for homoglyphs dicts: coming soon...
Scripts for ViT training and Inference for Ancient Chinese: coming soon...

For the Ancient Chinese Character Dataset:

Meta data file: coming soon..
Ancient Chinese Character Images: We draw our character images from https://xiaoxue.iis.sinica.edu.tw/yanbian
Folder of visualization: coming soon..

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
utils		utils
README.md		README.md
all_commands.sh		all_commands.sh
japan_supplier_match.py		japan_supplier_match.py
synthetic_accuracy_easy_2_gt.py		synthetic_accuracy_easy_2_gt.py
synthetic_accuracy_easy_2_paddle.py		synthetic_accuracy_easy_2_paddle.py
synthetic_accuracy_paddle_2_gt.py		synthetic_accuracy_paddle_2_gt.py
synthetic_placename_match.py		synthetic_placename_match.py

Provide feedback